PREDICTIVE ANALYTICS CH. 9 EXAM QUESTIONS WITH CORRECT ANSWERS LATEST UPDATE 2026

PREDICTIVE ANALYTICS CH. 9 EXAM QUESTIONS WITH CORRECT ANSWERS LATEST UPDATE 2026 Which of the following are true about classification algorithms? - Answers [1] They are the most used techniques in predictive analytics. [2] Classification algorithms are not limited to use by any single industry. [3] It is a form of supervised learning in which we wish to predict the class of each record in the data. Gartner reported that the value of firms heavily using predictive analytics was _____ higher than the norm. - Answers 200 to 300 percent Classification is a form of _____. - Answers supervised learning Which of the following are correct examples of predictive analytics use in industry? - Answers [1] Amazon used classification algorithms to up-sell or cross-sell to its customers. [2] Whirlpool Corporation uses classification algorithms to predict which dishwashers coming down the assembly line are likely or unlikely to fail. [3] eHarmony uses classification algorithms to classify individuals as either compatible or not compatible with eachother. Classification models would be useful for which of the following situations? - Answers [1] Deciding to treat one medical patient differently than another. [2] Making a decision who to send a mailing to. The Tobin's q for information-centered firms using predictive analytics were _____ higher than the average. - Answers 400 to 500 percent Which of the following would be the benefits of using categorizations? - Answers [1] Firms sell more product. [2] Health care is made more robust. [3] Banks suffer fewer loan defaults. Whirlpool Corporation uses classification algorithms to predict which dishwashers coming down the assembly line are likely or unlikely to fail. This is an example of what type of analytics? - Answers Predictive A real life application of classification would be _____. - Answers [1] Early detection of sepsis in hospitals. [2] An analytics algorithm assigned a category based on arrests, shootings, and affiliations with gang members. If you were trying to identify an unknown insect as either a katydid or a grasshopper, what are some characteristics you would use to aid in your identification? - Answers [1] How long is the abdomen? [2] Does it have wings? [3] How long are the antennas? The most used of the data mining algorithms are in the _____ category. - Answers classification Why is it a good idea to first graph the results of your classification? - Answers [1] Graphically looking at results lets you visually see classification relationships. [2] Graphically looking at results helps you identify geographical neighbors which will help in further identification. What does the "k" in k-Nearest-Neighbor refer to? - Answers The number of nearest neighbors used in determining a category correctly. When using k-Nearest-Neighbor, the training data set is _____. - Answers a subset of the total data available Which of the following statements are true for validation data? - Answers [1] Only use data that was not part of the training data set. [2] It is much like the difference between "in-sample" and "out-of-sample" test statistics in standard forecasting. [3] It is sometimes called testing data. What does the training data set do? - Answers It trains the model to correctly classify the unknown by selecting closeness to the k-nearest-neighbors. Which of the following are reasons why you would partition a data set? - Answers [1] It is a method for ensuring that no effectiveness statistics are drawn from the data used to create the model. [2] It is standard practice to divide the data set into partitions using some random procedure. [3] It produces reliable measure of the effectiveness of a data mining tool. T/F The real test of a data mining algorithm will be the statistics calculated from the training data, not the test statistics on the validation data. - Answers False What is the first step when using the data mining technique? - Answers Partition the data set. Diagnostics statistics for an estimated model would include which of the following? - Answers [1] The classification matrix [2] The lift chart If a calculation shows the percentage of the records that the model has placed in the incorrect category, then this would be called the _____. - Answers misclassification rate What do you do before building a data mining model? - Answers Partition the data set. What measures the change in concentration of a particular class when the model is used to select a group from a portion of the general population? - Answers Lift